Collaborative and Efficient Fine-tuning: Leveraging Task Similarity
This addresses data scarcity for users fine-tuning foundation models, but it is incremental as it builds on existing parameter-efficient methods like LoRA.
The paper tackles data scarcity in fine-tuning foundation models by leveraging task similarity across multiple users, proposing CoLoRA which uses shared and personalized adapters to boost individual performance, with experiments showing significant improvements when trained with similar tasks.
Adaptability has been regarded as a central feature in the foundation models, enabling them to effectively acclimate to unseen downstream tasks. Parameter-efficient fine-tuning methods such as celebrated LoRA facilitate efficient adaptation of large foundation models using labeled, high-quality and generally scarce task data. To mitigate data scarcity in fine-tuning of foundation models, we propose to leverage task similarity across multiple downstream users. Intuitively, users with similar tasks must be able to assist each other in boosting the effective fine-tuning data size. We propose Collaborative Low-Rank Adaptation, or CoLoRA, which exploits task similarity to collaboratively and efficiently fine-tune personalized foundation models. The main idea in CoLoRA is to train one shared adapter capturing underlying task similarities across all tasks, and personalized adapters tailored to user-specific tasks. We theoretically study CoLoRA on heterogeneous linear regression and provide provable guarantees for ground truth recovery. We also conduct several natural language experiments with varying task similarity, which further demonstrate that when trained together with similar tasks, individual performances are significantly boosted.